This post defends Functional Decision Theory (FDT) against criticisms, particularly those raised by Bentham's Bulldog on LessWrong. The author argues that FDT, when applied to scenarios like Newcomb's Problem, correctly identifies the optimal strategy by considering the implications of its own decision-making process. The author also addresses a more complex scenario, Newcomb's Revenge, demonstrating how FDT adapts its output based on the specific problem parameters and the agent's knowledge of the situation. AI
IMPACT Discusses theoretical decision-making frameworks relevant to AI alignment and agent behavior.
RANK_REASON The article is a philosophical defense of a decision theory, not a new release or significant industry event.
- Bentham's Bulldog
- Functional Decision Theory
- LessWrong
- Newcomb's Problem
- Newcomb's Revenge
- Omega
- Omicron
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
